Supervised Machine Learning Classifier for Email Spam Filtering

Abstract

Email is the omnipresent and persistent consequence applied to a regular base by a huge number of individuals worldwide. However, as a result of social support systems and advertisers, the majority of the emails hold unnecessary information called spam. This concern not just affects typical users of the net, but additionally causes an enormous setback for companies and organizations as it costs a massive amount of money in mislaid productivity, wastage of user’s time, and network bandwidth. In recent times, various parallel researchers have presented several email spam classification techniques, but it is extremely tough to eradicate the spam emails completely, while the spammers transform their techniques frequently. The proposed method is an efficient technique to classify the email spam messages using Support Vector Machines (SVM). Here, we present an SVM handling separation of nonlinear data using a Kernel function, which is an advanced machine learning technique in R to improve the accuracy of the model. Finally, we present a generic template for a working of kernel function in SVM that can be built in R.